Project 4 (4 Seasons alpha)

I had a bit of an issue with coming up with a good idea for the randomly generated project, so I ended up coming up with my final project idea first and then backtracking it as my project 4. While lighting and appropriate terrain design were large parts of my focus, the key to the aesthetic (RWBY) I wanted to capture relied on coding a convincing petal drift behavior, as well as its random placement and distribution on the randomly generated tree positions.

My first finished build of the petal/tree generation ended with about 1500 petals. I was genuinely surprised that the game was perfectly runnable, but I feel like that stylistic choice to make the petals unlit color textures actually helped ease the need to render reflections and shadows. I ended up creating different distributions for each tree, as my decided limit/range of 50-300 petals per tree wasn’t enough for the small thin trees to look convincing (they just looked like they had a loose ball of red particles floating above them).

Coding the actual drifting behavior was difficult, but in a different way than the tree/petal generation, which was mainly just large, complex, and took a lot of lines. The drifting behavior is technically only 8 lines or so; however, it took me a really long time to find good exact numbers for each rotational direction, as well as speed ranges for each movement axis, for the petals to actually feel like petals, and with a satisfyingly stylistic flow of movement. I ended up using a function called deltaTime, which is a measure of the time in between frames. Trying anything else, even the lowest possible float of 0.01, wasn’t small enough for the petals to actually drift, and not blast off into the horizon as if shot from a confetti cannon.

I’m still far from my original ideal, as I wanted the petals to have actual easing in their movement, where they dip and slow, and fall in a pendulum-like swinging rhythm. In the current build, the petals simply move in a certain direction for a random amount of time before changing directions, making hard angles and abrupt changes in speed. I actually abandoned my original plan for springtime, which would’ve been a lone tree on a vibrant green grassy hill with a bright blue sky and puffy white clouds, as it relied heavily on this lazy drifting behavior. Honestly though, I ended up actually really liking the way these petals drift in a purposeful direction alongside the golden cube while you’re running along the bone white structure (which I came up with pretty late into the actual final project development). I still definitely want to refine the drifting code; there’s a lot of aesthetic potentials I can feel that I’m nowhere close to yet that I want to be able to use.